EURIPIDES AND THE'TALES FROM EURIPIDES':
SOURCES OF APOLLODOROS''BIBLIOTHECA'?
In their search for the arguments of lost Euripidean tragediesscholars have often taken recourse to the mythographic manualsfrom Roman times, the Fabulae of Hyginus and the Bibliotheca ofApollodoros. However, the quality (direct or indirect) and therelative importance of the dependence of these mythographers onthe tragic poet, who lived some five or six centuries before l ), havenever been elucidated. The question has been complicated by thediscovery of several pafyrus-fragments of an alphabetic collectionof tragic hypotheses 0 Euripides, the so-called Tales from Euripides, a book of which the existence had already been surmised byWilamowitz2). Ever since, it has been stated again and again that
1) The date of the Library remains uncertain and can vary between 50 B. C.and 250 A. D. Most authors thought of the second century A. D. (see for exampleC. Robert, De Apollodori Bibliotheca, diss., Berolini 1873, 38 H.; M.-M. Mactoux,Pantheon et discours mythologique. Le cas d'Apollodore, RHR 206 [1989] 247),but G. Zuntz, The Political Plays of Euripides, Oxford 1955, 138-139 preferred thefirst century B. c., M. Van der Valk, On Apollodori Bibliotheca, REG 71 (1958)167 the first century A. D., and ]. C. Carriere, B. Massonie, La Bibliothequed'Apollodore, traduite, annotee et commentee (Lire les polytheismes, 3), Paris1991, 11 even date the work to the end of that century or the beginning of thefollowing on the ground of Pausanias' obvious ignorance of the work.
2) U. von Wilamowitz-Moellendorff, Analeeta Euripidea, Berlin 1875,182-184. Ever since, many separate papyrus-fragments have been published: seethe surveys in O. Bouquiaux-Simon, P. Mertens, Les temoignages papyrologiquesd'Euripide: liste sommaire arretee au 1/6/1990 in: Papiri letterari greci e latini, A
Euripides and the 'Tales from Euripides' 309
both mythographers relied to a great extent or even exdusively onthis collection for their knowledge of Euripides' plots. In view ofthe importance of these texts for the reconstruction of fragmentarytragedies, it is worthwile to examine the specific question whetherthey relied directly on these plays or on hypotheses or on otherintermediary sourees. In this paper I will focus only on Apollodoros: as to Hyginus, the question is obscured even more by theheterogeneous and poorly transmitted text and the fact that theGreek original of this Latin translation is lost.
There is much uncertainty about Apollodoros' sources ingeneral: the mythographer explicitly quotes Hesiod, Homer,Pherekydes, Akousilaos as well as some other minor mythographers and poets of the epic cyde, but also Euripides (fourtimes) and the tragedians (three times)3). However, the authorfollows other sources as well, without naming them, such as Hellanikos, used by Apollodoros for his presentation of Trojan history4). It is dear, indeed, that the explicit references, oftengrouped in "Zitatennester", always deal with specific variants ordeviations from the version currently followed by the mythographer. Sometimes these variants concern only proper names orother details. The question then arises whether the author borrowed, on the one hand his main argument, and on the other thevariants, from the original works, or from later epitomes, hypotheses, learned commentaries or a mythographical manual ormanuals. Although C. Roben, the first scholar to study the question in detail, thought that Apollodoros had consulted Pherekydes, Akousilaos and Asklepiades directly, he defended his dependence on intermediary sourees, mainly scholiasts and hypotheses,for his knowledge of other authors, such as Homer and the tragedians5). However, most scholars of the late 19th century and
cura di M. Capasso, Lecce 1992, 106-107; H. Van Looy, Les fragments d'Euripide.Premiere partie: ALyEuc; - 'htltOAU'tOc; KUAUltt0flEVOc;, AC 61 (1992) 297-299 (withbibliography on each hypothesis). One should add now the recently publishedfragments of a hypothesis of the Hippolytos Kalyptomenos (cf. W. Luppe, DieHypothesis zum ersten 'Hippolytos' (P. Mich. inv. 6222A), ZPE 102 [1994] 23-39),and of the Bacchae (P. Oxy. 60.4017).
3) Carriere, Massonie (cf. supra n.l) 12-17; M.Rodriguez de Sepulveda,Apolodoro, Biblioteca, Introd. deJ. Arce, Madrid 1985,22-28; C.Jourdain-Annequin, Heracles aux portes du soir. Mythe et histoire (Centre de recherches del'universite de Besanc;on, 89), Paris 1989, 235-242.
4) Cf. Van der Valk (cf. supra n.l) 134-143.5) Cf. Robert, De Apollodori ... (cf. supra n.l) 55; C.Robert, Bild und
Lied, Berlin 1881, 242ff.
310 Marc Huys
early 20th century, such as Wilamowitz, E. Schwanz, E. Bethe,W. Radtke, R. Wagner and A. Söder6), defended a variant of thethesis of the intermediary sources. Only for some specific passages, such as ApolIod. 3.5.7 H. [48 ff.] on the Oidipous-Iegend,did Schwanz and Roben argue a direct dependence on Greektragedy, in casu the prologue of Euripides' Phoinissai and extractsfrom Sophokles' Oidipous Tyrannos, a theory which was criticizedby Bethe7
). But in more recent scholarship there is a tendency toreassess the author as one who had consulted the original sources,and to appreciate his erudition and acquaintance with the originalachievements of dassical, often even archaic literature8), the predilection for the epic cyde and the older mythographers is indeednotewonhy and may reflect the archaizing fashion of the SecondSophistic.
Instead of going into this question in general, the purpose ofthis paper is to analyze the limited number of passages in whichApollodoros refers to Euripides or else relies on hirn without ex-
6) Wilamowitz (cf. supra n.2); E. Schwartz, De scholiis ad historiamfabularem pertinentibus (Neue Philologische Jahrbücher. Suppl.-Bd. 12), 1881,405(already suggested the dependence of both Apollodoros and the comparable [0'togLm in the Homeric scholia on several mythographic compendia from the Hellenistic period); M. Wellmann, De Istro Callimachio, diss., Gryphiswaldiae 1886,44-70 (tried to show that the klJVUYWyTj 'tWV 'A't8LÖWV of Kallimachos' discipleIstros was used by Apollodoros as a source of his chapters on the Attic mythological past); E. Bethe, Quaestiones Diodoreae mythographae, diss., Gottingae 1887(defended the hypothesis of a mythographic manual from the first half of the firstcentury B. C. as a common source for Diodoros, Hyginus and Apollodoros);W.Radtke, De Lysimacho Alexandrino, diss., Strassburg 1893, esp. 102ff. (theN60'tOL and 0TJßm'Kwv nuguö61;wv OlJvuyWyTJ of this historian from the fourth orthird century B. C. are considered an important source for Apollodoros or themythographic compendium on which he depended); R. Wagner, Epitoma Vaticanaex Apollodori Bibliotheca. Accedunt curae mythographae de Apollodori bibli?thecae fontibus, Leipzig 1.891 (defended the thesis that Apollodoros had used akind of summanes of the EplC cyde for his treatment of the TroJan saga); A. Scider,Quellenuntersuchung zum 1. Buch der Apollodorschen Bibliothek, Würzburg1939, esp. 166-169 (conduded that Apollodoros used prose summaries of epicpoems, hypotheses, commentaries etc. but never the original texts). See alsoC. Lütke, Pherecydea, diss., Gottingae 1893, 32 ff., who showed at least for theHerakles-Iegend that Pherekydes cannot have been a direct source for Apollodoros.
7) C. Robert, Oedipus. Geschichte eines poetischen Stoffs im griechischenAltertum, Berlin 1915, I 544-546; Schwartz (cf. supra n. 6) 450 ff.; Bethe (cf. supran.6) 85-86. .
8) Already Van der Valk, who adopted the theory of the mythographicalmanual as a source only for specific parts of the Library, such as the life story ofHerades (cf. supra n.1, 146ff.), but certainly Rodriguez de Sepulveda (cf. supran.3) 27 and Carriere, Massonie (cf. supra n. 1) 11 ff.
Euripides and the 'Tales from Euripides' 311
plicit reference. This specific case-study may throw new light alsoon the general question of Apollodoros' sources. I deal first withthe explicit references to tragedies of Euripides9).
(2.1.4 = 11) Bi'jAoe; öE Vn0I-lElvue; ev Alyu:1ttcp ßumAEun I-lEVAlyu:1tto'U, YUI-lEL öE 'AYXLVOl1V 'tTjv NElAO'U S'UYaLEQU, xui ulmlJ YlVOV'tm nuIÖEe; Ö(Ö'UI-lOL, Al:Y'U:1ttoe; xui ßuvuoe;, we; M CjJ110LV EUQLnlÖl1e;, xuiKl1CjJEiJe; xui <l>LvEiJe; nQooELL.
That Belos also fathered Kepheus and Phineus is nowherementioned in extant Euripidean tragedy, but it probably refers tothe Andromeda, where the genealogy of the heroine, Kepheus'daughter, is likely to have been given in the prologue 10). Somewhatsurprisingly, however, schol. E. Supp. 318 informs us that 0 EuQL:1t(Öl1e; CjJl10i nEV'tE nuIÖue; Elvm BijAO'U, Al:y'UnLOv, ßuvuov, <l>O(VLXU,<l>LvEU, 'Ayi]voQu. H.]. Mette suggested that the scholiast had erroneously omitted Kl1CjJEU here and that he in fact was referring tothe same Andromeda-passage as Apollodoros, but F. Bubel hasdenied this quite convincinglyll). The explicitation that Belos hadfive sons is indeed an additional difference, and the Phoinix ofEuripides is certainly another possibility. EIsewhere, the versionaccording to which Kepheus is the son of Belos seems to occuronly in Hdt. 7.61 (nuQa Kl1CjJEU 'tov BijAO'U); on the other hand,schol. in Arat. 179 calls Kepheus Belos' grandson I2).
(3.6.8 = 74-75) 'l0l-luQoe; I-lEV yaQ '!nn0I-lEÖOVLa unEx'tnVE,AECtöl1e; ÖE 'E'tEOXAOV, 'AI-lCjJ(ÖLXOe; ÖE TIuQSevonuIov. we; öE EUQLnlÖl1e;CjJl10l, TIuQSEvonuIov 0 TIoonöwvoe; nuIe; TIEQLXAUI-lEVOe; unEx'tELvE.MEAaVL:1tnOe; ÖE 0 AOLnOe; 'tWV 'AoLaxOÜ nu(Öwv Ele; 'tTjv yuO'tEQU T'UMu'tL'tQwoxn.
The reference to Parthenopaios' death by the hands of Periklymenos agrees with a detail told in the messenger speech ofEuripides Phoenician Women (1153-7), where a similar indicationof his descent is added: EVUA(O'U SEOÜ TIEQLXAUI-lEVOe; nuIe;. It is clearthat Apollodoros only mentions this as a particular variant sincethe remaining 'catalogue of killings' does not derive from
9) All extracts from the Library quoted are taken from Mythographi Graeci.I. Apollodorus. Pediasimus, ed. R. Wagner, Leipzig 21926.
10) Cf. F. Bube!, Euripides' Andromeda (Palingenesia 34), Stuttgart 1991, 24.11) H. J. Mette, Euripides (insbes. für die Jahre 1939-1968), 1. Hauptteil:
Die Bruchstücke, Lustrum 12 (1967) 57 (F "208); Bube! (cf. supra n.l0) 19.12) On Kepheus' descent, see Preller, Robert, Griechische Mythologie, II 1,
Berlin 1920, 237.
312 Marc Huys
Euripides. The variant, however, is not exclusively Euripidean:Paus. 9.18.6 informs us that Asphodikos killed Parthenopaios according to the Thebans 13), but that the Thebaid reserved the featfor Periklymenos, and from a scholion to Ph. 1156 we know thatthe killing was also recorded by Aristodemos, who specified thatPeriklymenos struck Parthenopaios with a stone14).
(3.7.7 = 94-95) EUQl:1tLÖT]~ OE <jJT]<JLV 'AAxl!aLWVa xanl tOV tfj~
l!avLa~ XQavov EX Mavtoii~ TELQWLO'U :rta'[öa~ Mo yevvfj<Jm, 'AI!<jJLAOXOV xai S'UyatEQa TWL<jJaVT]V, XOI!L<JaVta ÖE EL~ KaQLVSOV tU ßQE<jJT]Öoiivm tQE<jJELV KOQLVSLWV ßa<JLAE'[ KQEOVtL, xai tT]v I!EV TWL<jJaVT]VÖLEvEyxoii<Jav EUI!OQ<jJLC;X uno tfj~ KQEOVto~ y'UVmxo~ UnEIJ.3tOAT]Sfjvm,ÖEöOLx'ULa~ I!T] KQEWV aUtT]v yal!EtT]V noL1]<JT]tm. tOV öE 'AAxl!aLWVauyoQu<Javta taVtTlV EXELV oux ELöOta tT]v ta'Utoii S'UyatEQaSEQunmvav, naQaYEVal!EVOV öE EL~ KaQLVSOV Eni tijv tmv tEXVWVunaLtT]<JLv xai tOV 'Utov XOI!L<Ja<JSm. Kai AI!<jJLAOXO~ xatu XQT]<Jl!oiJ~
,AnaAAWVO~ ,AI!<jJLAOXLXOV "AQYo~ Q>XWEV.Scholars unanimously agree that this passage, added by Apol
lodoros as an appendix to the legend of Alcmeon, reproduces theargument of Euripides' posthumous Alcmeon at Corinth 15). Thelegend indeed seems to have been invented to a large extent byEuripides hirnself and is not further attested apart from this Apollodorean addition. It is striking, however, that the action of theplay is covered only by the clause naQaYEVal!EVOV ÖE EL~ KaQLVSOV
13) ,AOlpO/'>LXOC; was corrected by Dindorf to 'AIJ.q>LÖLXOC; on the basis of thepassage quoted from the Library, but 'AOq>OÖLXOC; is attested as a proper name inHerodianus 3.1, p.152.7, 3.2, p.445.3, 3.2,/.926.20. More probably, the Apollodorean 'AIJ.q>LÖLXOC; should be correcte to'AOq>OÖLXOC;, as suggested byWilamowitz: cf. F. Vian, Les origines de Thebes. Cadmos et les Spartes (Etudes etcommentaires 48), Paris 1963, 203. According to the collation made byM. Papathomopoulos, Pour une nouvelle edition de la Bibliotheque d'Apollodore,26 (1973) 22, the manuscript R has 'AIJ.q>OöLXOC;.
14) Cf. Radtke (cf. supra n.6) 99. R. Klotz, Euripidis Phoenissae, rec. etcommentariis instr., Berlin 1841, 181 and C. Robert, Griechische Mythologie, II.Die griechische Heldensage 3.1, Berlin 1921,940, were convinced that Euripidesderived his version from the Thebaid. See for the fragment of Aristodemos, probably taken from the 'EmYQulJ.lJ.am 0T]ßaLxu of Aristodernos of Thebes, the pupilof Aristarch, also W. Radtke, Aristodems 'EmYQulJ.lJ.a1:a 0T]ßmxu, Hermes 36(1901) 49.
15) F.A. Basedow, De Euripidis fabula quae inscribitur 'AA.XIJ.ElllV Ö öu].KOQLv801J, Berlin 1872; Th. Zielinski, De Alcmeonis Corinthii fabula Euripidea,Mnemosyne 50 (1922) 305-327; T. B. L. Webster, The Tragedies of Euripides,London 1967,265; H. Van Looy, Zes verloren tragedies van Euripides (Verhandelingen van de Koninklijke Vlaamse Akademie voor wetenschappen, letteren enschone kunsten van Belgie 25.51), BrusseI1964, 103.
Euripides and the 'Tales from Euripides' 313
btL 'tT]V 't(tlV 'tEXVWV UnULUjOLV XUL 'tov vEov XO!tLouoOm (UVUXO!tLOUOOm corr. Hercher), whereas the preceding text exclusively dealswith the nQoYEYEVTj!tEVU and the last sentence contains a prophecyprobably spoken by Apollo as a deus ex machina I6). It seems, then,that Apollodoros' source was an ancient hypothesis of the play butthat the mythographer omitted the summary of the actual intrigue;even the recognition of Alcmeon's son and daughter is not mentioned 17). The text indeed contains some typical characteristics ofthe Tales from Euripides: it begins with the proper name of thetitle-hero ('A'AX!tULWVU); the use of the participle nUQuYEv0!tEVO<; toindicate the first entrance of a hero on stage (hyp. Alex., Andr.,Hec.)18); the preference for balanced !tEv-öE-constructions ('tT]V !tEVT ' , ö' 'A" )19)LOLqJOVTjV ... , 'tOV E ",X!tULWVU . . . .
(3.9.2 = 109) 'HOLOÖO<; öE XUL nVE<; E'tEQOL 'tT]V 'A'tUAUV'tTjV oux'Iuoov uAAa LXOLVEW<; Einov, EUQLnLÖTj<; ÖE MmvuAov, xUL 'tov Yl1!tuv'tU uu'tT]v ou MEAUVLWVU uAAa 'Inno!tEvTjv. EYEVVTjOE ÖE EX MEAUVLWVO<;'A'tUAUV'tTj f\ "AQEO<; lluQOEvonulov, ö<; EnL 8i]ßu<; EO'tQU'tEUOU'tO.
Apollodoros clearly means here that Euripides differred fromall other authors in calling Atalante the daughter, not of Schoineusor lasos, but of Mainalos - the reference to the varying versions ofthe heroine's husband should not be related to Euripides but to'Hesiod and some others'20). He refers to E. Ph. 1162: 'tu XUAAL'tO-
16) The text of Apollodoros suggests that Amphilochikon Argos was founded by Alcmeon's son, whereas Thuc. 2.68.3 specifies that this Amphilochos wasAlcmeon's brother: so Euripides must have followed another version, unless Apollodoros' account is inaccurate here - Carriere, Massonie (cf. supra n. 1) 229 wrongIy suggest that it may be an interpolation.
17) Carriere, Massonie (cf. supra n. 1) 229 therefore suggest a lacuna afterrutaL-tTlOLv to be filled out with avayvwQLom or avayvwQLoao8m, but this seemsartificia!.
18) Cf. J. Krenn, Interpretationen zu den Hypothesen in denEuripideshandschriften, diss., Graz 1971, 17, 191; W.Luppe, Die Hypothesis zuEuripides' 'Alexandros', Philologus 120 (1976) 14 (hyp. E., Alex. 30); W. Luppe,Zur 'Alexandros'-Hypothesis (P. Oxy. 3650), ZPE 63 (1986) 7-8.
19) Cf. J. Diggle, The Papyrus Hypothesis of Euripides' Orestes (P. Oxy.2455 frA co!. IV 32-9 + fr. 141), ZPE 77 (1989) 3-5.
20) lt cannot be doubted that most modern translations are wrong: J. G.Frazer, Apollodorus. The Library (Loeb Classical Library), 1921, I 403; K. AIdrich, Apollodorus, The Library of Greek Mythology, trans!. with notes and indices, Lawrence, Kansas 1975; Rodriguez de Sepulveda (cf. supra n. 3) 166; Carriere,Massonie (cf. supra n. 1) 102 ("Euripide en fait la fille de Mainalos et lui donne pourepoux Hippomenes, et non Melanion"). Frazer even thought that the names of thefather and husband of Atalante figured in the lost Meleagros, ignoring the referenceto the Phoinissai. But it is known from fr. 74 M.-W. from the fuvmxwv
314 Marc Huys
1;q> 1-t'l]l:Q~ MmVUAOlJ x6QU: this is a periphrasis for Atalante, themother of Parthenopaios. Ever since the time of the scholiasts,there has been disagreement among critics as to whether MmVUAOlJx6Q'I] should be interpreted as 'the daughter of Mainalos' or as 'thegirl of (i. e. staying on, roaming on) Mount Mainalos'. Mainalos,like Lykaon, was indeed known in antiquity both as name of amountain and as a personal name21 ). The ancient scholion on thePhoinissai-line prefers the latter possibility: l:TI uno MmVUAOlJoQOlJS; 'AQxaö[as;, but mentions the alternative reading MmVUAOlJx86va, probably a conjecture that must have resulted from theuncertain interpretation of the line. I think, however, that Apollodoros is right and that Scherling justly explained the preferenceof the ancient scholiast: "Doch scheint dies nur eine Vermutungdes Scholiasten zu sein, um die Ungleichheit in den Angaben überden Vater zu beseitigen". Elsewhere in Euripidean tragedy x6Q'I]with the genitive of a proper name always means "the daughterof ... " (cf. Med. 1234: x6Q'I] KQEOVWS;; Hipp. 15, 713: "AQl:El-tLV,~LOS; x6Q'I]v; Andr. 897, 1049: MEVEAEW x6Q'I](v); Ba. 2: KuÖl-t0lJx6Q'I]; IA 701: N'I]QEws;x6Q'I]v, 1444, 1543: ~Los;x6Q'I]S;). Moreover, itis known from schol. Apoll. Rh. 1.769 that Hellanikos, a contemporary of Euripides, already connected the figure of Atalante notonly with the mountain, but also with the hero Mainalos (=FGrHist 4 F 162): Ma[vaAos; OE oQOS; 'AQxaö[as;, f:v <P f) 'AWAUVl:'I]ÖL'fjyEv, uno MmVUAOlJ wü 'AQxuöoS;, WS; cprjOLV 'EAA.aVLXOS;22).
It is striking that two of the four references to Euripidesconcern the same passage from the Phoinissai about the killing ofthe hero Parthenopaios. In either case Euripides is referred to for a
Ka1;aAoyo~ that Hippomenes was connected with Atalante in this work: cf.J. Schwartz, Pseudo-Hesiodeia, diss., Paris 1960, 129.
21) Cf. Scherling in RE XIV 1, Stuttgart 1928, col!. 576-578, s. v. Mainalos(-on). See ApolIod. 3.8.1.3 and scho!. Apoi!. Rh. 1.168 (for other references seeFGrHist I a: Nachträge, Kommentar, ed. F.Jacoby, Leiden 1957,470). Paus. 8.3.4calls Mainalos the founder of the city Mainalos and Paus. 8.36.7-8 describesMainalos as an Arcadian mountain and a plain as weil as a city.
22) Mainalos as a personal name is accepted by L. C. Valckenaer, Euripidistragoedia Phoenissae, Lugduni Batavorum 1802, 369; Klotz (cf. supra n. 14) 182;Preller, Robert (cf. surra n.12) 94; Scherling (cf. supra n. 21) co!. 578, but explicitlyrejected by Jacoby (c . supra n. 21) 470; E. Craik, Phoenician Women, edited withtranslation and commentary, Warminster 1988, 236; J.M. Grove, Euripides.Phoinissai, van inleiding, tekskritiese apparaat en kommentaar voorsien, PortElisabet 1984,140; Euripides. Phoenissae, ed. with intro and comm. by D.J. Mastronarde, Cambridge 1994, 474. The same controversy exists in connection withStatius, Theb. 6.541 Maenaliae Atalantes in a line criticized by the scholiast anddeleted by some scholars.
Euripides and the 'Tales from Euripides' 315
learned variant that is also treated in the scholia. Very probably,then, both references go back to Hellenistic scholarship, perhapsDidymos, from which also the mythographic D-scholia derive23).
Concerning the Atalante-passage, J. Schwartz already concluded24): "c'est sans doute aun commentaire de tragedie que nousdevons ce fragment." Also in the first passage, the reference toEuripides concerns a genealogical detail that seems borrowed fromsome learned commentary rather than from the reading of theoriginal tragedy.
In the following three passages Apollodoros does not refer toEuripides personally but to the tragedians in general:
(2.1.3 = 5) "AQyo1! ÖE xui '10I-lf]vTjpfj\; 'Aownoii nUL\; 'luoo\;, oi"!CPUOLV 'lw yEvEo8m. KaenwQ ÖE 6 o1!YYQa'IjJu\; 'ta XQOVLXa xui nOAAoi't(DV 'tQUYLXWV 'Ivaxo1! 'tTjv 'lw My01!OLV' 'HOLOÖO\; ÖE xui 'AX01!OLAUO\;IIELQTjvo\; UtJLf]v cpUOLV dvm.
The vague nOAAoi 'twv 'tQUYLXWV certainly points to Aischylos(Prom. 589), Sophokles (EI. 4 and probably Inachus F 269 d 23,284 Radt) and Euripides (Supp. 629: LW ZEii, 'tä\; nUAm0l-laLOQO\;nmöoyavE naQLO\; 'Ivaxo1!), but possibly also to a play by a minortragedian, such as Chaeremon's /0 25 ). The vagueness of the reference to the tragic texts and its insertion in a chain of references26)
make it again very doubtful that Apollodoros would have consulted here the tragic passages themselves: rather he used learned commentaries or previous mythographers.
(2.1.5 = 23) 'Al-lwwvTj öE EX IIoOELÖWVO\; EyEvvTjoE NuvnALOv.oi"!'to\; I-lUxQaßLO\; yEVaI-lEVO\;, nMwv 'tTjv 8aAuoouv, 'toL\; E!!l'tLl'tL01!OLV
23) On the link between these scholia and Apollodoros, see Wagner,Mythograrhi Graeci. I ... (cf. supra n.9) XXXIII-XXXVI; A. Diller, The TextHistory 0 the Bibliotheca of Pseudo-Apollodorus, TAPA 66 (1935) 297-300; Vander Valk (cf. supra n.l) 143-152, and on some of their predecessors found onpapyrus: F. Momanari, Studi di filologia omerica amica I (Biblioteca di studi amichi 19), Pisa 1979, 14-15.
24) J. Schwartz (cf. supra n. 20) 129.25) On the play of Chairemon, see C. Collard, On the Tragedian Chaere
mon, JHS 90 (1970) 26. It is striking that modern scholars ·surveying the differemgenealogies of 10 ignore the Euripidean passage: cf. Engelmann, in: AusführlichesLexikon der griechischen und römischen Mythologie, hrsg. v. W. H. Roscher, II.1,Leipzig 1890-1894, co1.125; Eitrem in RE IX 2, Stuttgart 1916, s. v. 10, co1.1732;Fr. Stoessl, Der Prometheus des Aischylos als geistesgeschichtliches und theatergeschichtliches Phänomen (Palingenesia 24), Stuttgart 1988, 56.
26) qJUOLV is a correction of B. Aegius for qJT]oiv, but one might also think ofan omitted author's name: cf. Carriere, Massonie (cf. supra n.l) 177.
316 Mare Huys
€:rd 8avutQl €l't'UQoocpaQfL. ~'UVEßTl ouv xai autov tfAf'UtTjOaL €xdvQl t<TJ8avutQl qJl'tfQ aAAWv tfAf'UtTlOUVtWV HO'UocpaQfL, l'tQiv tfAf'UtTjOaL.'EYTll-Lf OE w~ I-LEV OL tQaYLxoi AEY0'UGL, KA'UI-LEVTlV tilV KatQEW~, w~ OE 6toiJ~ vaoto'U~ YQu\jJa~, <l>LA.lJQav, w~ OE KEQXW\jJ, 'HOLavTlv, xai eYEvVTlOf IIaAal-Lf]0TlV otaxa Na'UGLI-LEOoVW.
Again the reference to the tragedians is embedded in aseriesof quotations. But this time the tragic texts to which it refers arelost: we only know that Nauplios was the title-hero of plays byAischylos, Sophokles, Philokles I (24), Astydamas II (60 F 5) andLykophron (100 F 4a) and that tragedies were composed on hisson Palamedes by the three great tragedians, as weIl as by Astydamas II (60 F 5a) and Theodotos (157). Apart from this,Palamedes also figured in Euripides' Philoktetes where at least hisfather's name was mentioned according to the paraphrase in D.ehr. 59.8. The descendence of Palamedes from Nauplios and Klymene is also mentioned in Apollod. 3.2.2 [15] and Epit. VI 8, butotherwise it is only known from schol. E. Or. 432 (NmmALo'U xaiKA'UI-LEVTl~ tTj~ 'AtQEW~ [cod. M] eYEvOVto mal; xui IIuAaflf]OTl~). Itwas suggested by R. Wagner that the Apollodorean account (seealso Epit. III 8) depended on Euripides' Palamedes27), whereas,according to C. Robert, followed by F. Stoessl, F. Jouan, R. Aelionand C. W. Müller, the Orestes-scholion would summarize a hypothesis of this play28). However, T. B. L. Webster, and especiallyR. Scodel29), who analyzed this question most fuHy, refute this:Scodel derives the Euripidean version from Hyg., F. 105, wherethe lineage of Nauplios is lacking.
(2.2.1 = 25) xai xQatf]ou~ ,AXQLOLO~ IIQoltov "AQYo'U~el;fAaVVfL.6 0' ~XfV d~ A'UxLav l'tQo~ 'IoßutTlv, w~ OE tLVE~ cpaGL, l'tQo~ 'AI-LCPLUVUXw· xai yal-LfI tilv tOVtOU 8UYatEQU, w~ I-LEV "OflTlQO~, "AVtfLUV, w~ OE OLtQUYLXOL, ~8fvEßOLav.
27) Wagner, Epitoma Vatieana ... Curae Mythographae ... (cf. supra n. 6)180. However, a possible correspondence between the Palamedes-fabula of Hyginus and the tragedy of Euripides is not excluded.
28) Roben (cf. supra n. 14) 1133; F. Stoessl, Die Palamedestragödien der dreigroßen Tragiker und das Problem der Hypotheseis, WS 74 (1966) 100-101;F.]ouan, Euripide et les legendes des Chants Cypriens. Des origines de la guerre deTroie a I'Iliade (Collection d'etudes anciennes), Paris 1966, 345; R.Aelion,Euripide heritier d'Eschyle (Colleetion d'etudes anciennes) I, Paris 1983, 48-53with funher bibliography on p. 51 n.23; C. W. Müller, Der Palamedesmythos im'Philoktet' des Euripides, RhM 133 (1990) 203-204.
29) R. Scodel, The Trojan Trilogy of Euripides (Hypomnemata 60),Göttingen 1980, 47-54; Webster (cf. supra n.15) 175.
Euripides and the 'Tales from Euripides' 317
Stheneboia figured both in Euripides' homonymous play andin his Bellerophontes. In this case Apollodoros needed no detailedknowledge of Euripides' work, since the heroine figured in thetide of one of his plays and the marriage of Proitos withStheneboia was recorded in the hypothesis of that play: IIQoLtoc;... yfuwc; Öf ~efVEßOLUV. Of the other tragic poets we know onlythat Astydamas II also wrote a Bellerophontes (TrGF 60 F 19) - asfar as we know, Sophokles and Aischylos did not treat this legend.
The three last jassages are taken from the two first paragraphs of the secon book. All three references to the tragediansprobably include Euripides, but none of them seems to derivefrom direct reading30). The swift succession of these referencespoints to a source only used in this part of the work31 ), perhaps acommentary - the reference OL tQuyuw( also occurs in some scholiaquoting mythographical variants (schol. n. 9.481a2; schol. Od.11.260). All the examples studied, then, confirm the impressionthat Apollodoros did not consult the tragic text of Euripides, butonly indirect sources deriving from the tradition of Hellenisticscholarship.
However, a number of passages without explicit referencehave also been interpreted as deriving from Euripides or at leastfrom hypotheses of his tragedies. Ever since Robert andWilamowitz it is commonly accepted that some short sequencesfrom the Library reflect Euripidean hypotheses 32), and a warning
30) Van der Valk (cf. supra n.l) 162 n.224 oversimplifies when he writesthat "BibI. thrice mentions the tragedians, while he then has in mind Sophocles(once) and Euripides (thrice)".
31) See also Robert (cf. supra n. 1) 60.32) Robert (cf. supra n.l) 83 [Apollod. 1.9.28]; U. von Wilamowitz, Analec
ta Euripidea, Berlin 1875, 183-184; Robert, Bild ... (cf. supra n. 5) 243 [Apollod.3.5.2]; Wagner (cf. supra n. 6) 199; R. Wagner, Proklos und Apollodoros, Jbb. f. cl.Phil. (1892) 242-244 [Epit. VI 26-27]; Frazer (cf. supra n. 20) I 18 + n. 4, 5 [ApolIod. 1.9.15, 1.9.28]; C. H. Moore, Notes on the Tragic Hypotheses, HSCP 12(1901) 297 [Apollod., Epit. III.5]; Söder (cf. supra n.6) 131-132, 161 [Apollod.1.9.15, 1.9.28]; Krenn (cf. supra n.18) 2, 84-85, 94-96, 111 + n.3, 116, 165, 167,189,202,208-209,214 n. 2,215; B. Zühlke, Euripides' Stheneboia, Philologus 105(1961) 8 n.3 [Apollod. 1.9.15,2.8.1,3.5.2 sq., 3.5.7, Epit. I 18 sq., III 21 sq., VI23-25, VI 26 sq.]; Zuntz (cf. supra n. 1) 136, 141 n.6, 145 [Apollod. 1.9.28; 3.5.2;3.13.8]; A. W. A. M. Bude, De hypotheseis der Griekse tragedies en komedies. Eenonderzoek naar de hypotheseis van Dicaearchus, diss., 's Gravenhage 1977,109-110; P.Carrara, Dicearco e l'hypothesis dei Reso, ZPE 90 (1992) 38-39: "Lacollezione dei Tales come tale, invece, ebba una discreta yoga nell'eta imperiale,come dimostrano ... i significativi echi e riuttilizzi da parte di mitografi (Bib-lioteca di Apollodoro )"; M. Haslam, The Authenticity of Euripides, Phoenissae1-2 and Sophocles, Electra 1, GRBS 16 (1975) 152 n. 10, 154: "the mythographers
318 Mare Huys
remark like that of Rusten is quite exceptionaP3): " ... not everynarrative of a myth which happens to resemble a Euripidean plot(e. g. BibI. 3.5.2 [36], 3.13.8 [174]) is necessarily derived directlyfrom a collection of hypotheses". The passages from Apollodoroswhich may be related to or depend on a Euripidean hypothesis arethe following:
1.9.1 [80-2] = hypo Phrixos A; 1.9.15 [105-6] = hypo Alk.;1.9.27 [144] = hypo Peliades; 1.9.28 [145-6] = hypo Med.; 2.3.1-2[30] = hypo Stheneboia; 2.4.1 [34-5] = hypo Danae34); 2.7.4[146-7] and 3.9.2-4 [103-4] = hYf' Auge; 2.8.1 [167-8] = hypoHer.; 2.8.4-5 [177-8] = hypo 0 a Temenos-drama (P. Oxy.27.2455 fr. 9, fr. 10)35); 3.5.2 [36] = hypo Ba.; 3.5.5 [42-4] = hypoAntiope36); 3.6.4 [64-5] = hypo Hyps.; 3.5.7-3.6.1 [48-58] and3.6.6-3.7.1 [68-78] = hypo Phoinissai; 3.12.5 [148-50] = hypoAlex.; 3.13.8 [174] = hypo Skyrioi; Epit. I 2-3 = hypo Skeiron;Epit. I 18 sq. = hypo Hipp.; Epit. III 5 = hypo HeL; Epit. III 27and V 8 = hypo Philoktetes; Epit. V 23 = hypo Tr.; Epit. VI 23-25= hypo EI.; Epit. VI 26-27 = hypo IT.
Apart from these parallels, other passages have been connected with lost hypotheses, such as 1.8.2-3 [65-71] = Meleagros;2.4.1-2 [36] = Diktys; 2.4.3 [43-4] = Andromeda; 2.4.8 [61-2] =
... show that our Euripidean hypotheses enjoyed popularity - were standard, even- in the early centuries of the Empire"; J. Rusten, Dicaearchus and the Tales fromEuripides, GRBS 23 (1982) 357 n.2 [ApolIod. 3.5.5]; W. Luppe, Euripides-Hypotheseis in den Hygin-Fabeln 'Antiope' und 'Ino'?, Philologus 128 (1984) 4H5[ApolIod. 3.5.5]; W. Luppe, Die Hypothesis zu Euripides' 'Skyrioi', Anagennesis 2(1982) 268 [ApolIod. 3.13.8]; W. Luppe, Hypothesis zum ersten Phrixos desEuripides, APF 32 (1986) 7-13 [ApolIod. 1.9.1]; W.Luppe, Zwei Hypotheseis zuEuripides-Dramen der Ternenossage (P. Oxy. 2455 fr.9 und fr. 10), Prometheus 13(1987) 194-197 [Apollod. 2.8.3-4].
33) Rusten (cf. supra n. 32) 361 n.21.34) Ir remains possible that this hypothesis is a Byzantine forgery, but
W. Luppe, Die Hypothesis zu Euripides' 'Danae', ZPE 87 (1991) 1-7 and Nochmals zur 'Danae'-Hypothesis, ZPE 95 (1993) 65-69 has convincingly demonstratedthat, except for some changes possibly of a later date, it belongs to the ancientTales-rype.
35) These rwo scraps of the hypotheses-roll must be interpreted as fragmentsof a tragedy on the Temenos-legend: cf. A. Harder, Euripides' Kresphontes andArchelaos, Leiden 1988, 279, 288; W. Luppe, Zwei Hypotheseis ... (cf. supra n. 32)193-203.
36) This hypothesis is lost, but has been reconstructed by W. Luppe,Euripides-Hypotheseis in den Hygin-Fabeln 'Antiope' und 'Ino'?, Philologus 128(1984) 41-59 on the basis of Hyg., F. 8, a fable specifically introduced as being anargument of Euripides' Antiope.
Euripides and the 'Tales from Euripides' 319
Alkmene; 2.5.11 [116-7] = Bousiris; 2.8.5 [180] = Kresphontes;3.1.3 [8-11] = Kretes; 3.3.1 [17-20] = Polyidos; 3.7.5 [89-90] =Alkmeon Psoph.; 3.13.8 [175] = Phoinix; 3.15.4-5 [201-4] =Erechtheus; Epit. I 5-6 = Theseus; Epit. II 3-7 = Oinomaos;Epit. III 8 = Palamedes; Epit. III 19-20 = Telephos; Epit. III21-22 = IA; Epit. III 30 = Protesilaos.
Usually, in the quoted passages, the narrative of Apollodorosto a greater or lesser degree resembles that of the correspondinghypothesis. Sometimes, similar accounts are interrupted by significantly different details. According to the Epitome, for example,Hermes' abduction of Helen to Egypt was the will of Zeus (x.m:CtßovA:rlOLv ß~6~), whereas the manuscript hypothesis of the Helencorrectly specifies that this happened 'HQa~ ßOllAfj. Yet, the variantof the Epitome may derive from a misunderstanding of E. Hel.44-48 or from E. EI. 128237). Likewise, the Apollodorean accountof the legend of Phrixos' sacrifice closely follows the fragments ofhypo Phrixos A (P. Oxy. 27.2455, XVI 221 ff.; P. Oxy. 52.3652, II16 ff.), but 'Ae6.!-la~, BOLWt(a~ ÖllVaOtEVWV is in striking contrastwith the hypothesis where Athamas is called the king of Thessaly note, however, that Orchomenos is the place of residence ofAthamas in hypo Phrixos B (P. Oxy. 27.2455, XIX 270). Similarvariations have been shown to exist between the Apollodoreanaccounts on Menoikeus, on Kadmos the dragon-slayer and thefoundation of Thebes and the corresponding episode of Euripides'Phoenician Wornen and its hypothesis by F. Vian38).
Generally speaking, the accounts in the Library are muchshorter than the hypotheses and concentrate on the events essentialfor the continuation of the legend. This narrative material is oftenfound at the beginning of the hypotheses, the part correspondingwith the exposition of the 3tQOYEYEv1']!-lEva in the prologue, whereasdetails of the Euripidean plot, usually summarized by the hypotheses, are omitted or reduced to a single event in the Library.We have already suggested this procedure in the case of the Alcrneon at Corinth, and another nice example is Apollod. 3.12.5[148-50], where the story of Alexander's birth, abandonment andgrowing up among the herdsmen corresponds to a greater or lesserdegree with the former half of the hypothesis of the Alexandros (P.
37) Euripides, Helena, hrsg. u. erkl. v. R. Kannicht, Bd. II: Kommentar,Heidelberg 1969, 30 n.12.
38) Cf. Vian (cf. supra n. 13) 28-29, 212. This author concluded that the textof Euripides could not have been Apollodoros' source; more probably he used acommentary on the play.
320 Mare Huys
Oxy. 52.3650, I 1-14)39), while only one short sentence (xai !-tE't' OUJtOAU LOU<; YOVEa<; aVEüQE) matches the whole remaining part of thehypothesis, which summarizes the actual plot of Euripides' play.On the other hand, Van der Valk40) has shown that Apollodorosgoes back to Hellanikos here, an author who seems to have beenused frequently by Euripides41 ). Even the short manuscript hypothesis of the Trojan Wornen mentions the debate scene betweenHelen and Hecuba ('Exaß'lJ bio 'tij<; !-tEv 'EAEv'lJ<; xu't'lJyoQftaaaa),which is probably a specifically Euripidean scene, whereas mostdetails of the legend common to the Apollodorean Epitome andthe hypothesis, such as the murder of Astyanax and the sacrifice ofPolyxena, are also found in other sources.
In some cases, it is by no means certain whether Apollodoros,despite obvious similarities, follows a Euripidean version. Epit. III27, for example, very probably reflects the tradition of the KUJtQLa,whereas V 8 seems to follow a post-Euripidean version, in whichHelenos and Machaon of Euripides' Philoktetes were replaced byKalchas and Podaleirios respectively42). Especially for the caseswhere there is no hypothesis left, the supposed correspondences
39) It remains a matter of debate how far the parallelism goes between thetwo texts. At any rate Hekabe's dream as the cause of child exposure (but notnecessarily the interpretation by Aisakos), the description of the act of abandonment and the name-giving by the herdsman are striking correspondences, althoughmuch depends on the interpretation of the first hypothesis-lines: see on this knottysubject R. A. Coles, A New Oxyrhynchus Papyrus: The Hypothesis of Euripides'A1exandros, BICS Supplement 32, London 1974, 17-18; M. Huys, Some Reflections on the Controversial Identity of the :7tQeoßu<; in Euripides' "Trojan Women"and in his"Alexander" (fr.43, co!. III, 12), AC 54 (1985) 245-246 with n.l0.Certainly the old theory that Apollodoros' account reproduces the contents ofEuripides' tragedy, whereas that of Hyginus, F.91 would summarize the contentsof the Alexandros of Sophokles (cf. G. Wentzel, F. Spiro, Epithalamion für Wolfgang und Helen Passow, Göttingen 1890, 54; 1. Rasch, Sophocles quid debeatHerodoto in rebus ad fabulas exornandas adhibitis, Jena-Leipzig 1912, 43 n. 2, 47n.2, 50-52) is outdated, just as the opposed thesis, defended by C. Lefke, DeEuripidis Alexandro, diss., Bochum 1936, 17-19, that Hyginus would derive fromthe Alexandros of Euripides and Apollodoros from the tragedy of Sophokles.
40) Van der Valk (cf. supra n. 1) 137-139.41) Cf. Aelion (cf. supra n.28) 210_n.60; FGrHist. 4 F 43: (concerning the
Alope-legend) ... 00<; 'EAAaVLXO<; TE tv ß 'AT8(öo<; xai EUQL1t(ÖTj<;; 4 F 96-99:Hellanikos has the same versions as those used by Euripides in his Phoinissai, 4 F126 (placed the residence of king Athamas in Orchomenos, just as Euripides in hisPhrixos B). See already on the common link between Atalante and Mount Mainalossupra p. 313 f.
42) Cf. Jouan (cf. supra n.28) 313-314; C. W. Müller, Patriotismus undVerweigerung. Eine Interpretation des euripideischen Philoktet, RhM 135 (1992)104-106 n. 2, 3, 4.
Euripides and the 'Tales from Euripides' 321
are often debatable, certainly for fragmentary plays. The Epitome's account on Telephos, for example, seems to derive in oneway or another from the KUTtQLa, whereas the connexion withEuripides' tragedy is uncertain43). Another example is the Epitome's version of the Polyidos-Iegend, which has been connectedwith Aischylos' Kressai as weIl as with Sophokles' Manteis andEuripides' PolyidoS44 ). And for the Meleagros-Iegend critics generally agree that Apollod. 1.8.2 [65-71] corresponds with the Euripidean version, but several details such as Meleagros' marriage withKleopatra are more probably borrowed from Homer, Sophoklesor even from other traditions45).
Looking subsequently at the formulation, one observes thatthe total number of verbal parallels between the relevant passagesin Apollodoros and the text of the hypotheses is not particularlyimpressive either:
(1) Apollod. 1.9.1 = 81: dC; LlEAepoiJC; ... aVETtEL<JE ... 'tl]vaxaQTt(av, Eav oepaYii LlLi 6 <l>QU;OC; = hypo Phrixos I: axaQTt(ac; [... ]AtlOLV,EL <l>QU;OC; [...] dc; LlEAepoiJC; aTt[... ]AOV ETtELOE ....
(2) 1.9.15 = 106: 'ATtoJ..J..wv ... l1'tf]OaLO TtaQa ~OLQWV = hypoAle.: 'ATtoJ..J..wv l1'tf]oa'to TtaQa 'twv ~OLQWV. Note that E;n'tf]Oa"CO isattested in the same mythical context in schol. E. Ale. 12.
(3) 1.9.28 = 145: "CO'll 'tijc; KOQ(vSotl ßaoLAEwc; KQEOV"COC; 'tl]v StlYaLEQa rJ..auxllv '!UOOVL EyytlWV'toC; = hypo Med.: '!UOWV ... EyytlU'taLxai Ll]V KQEOV"COC; "CO'll KOQLVS(WV ßaoLAEwc; StlYaLEQa rJ..auxllv. Theverb EYYUUW is not used in the Medeia of Euripides, but is ordinaryclassical Greek, both active ('to betroth': cf. E. IA 703) andmedium ('to become engaged to': cf. Dem. or. 57,41).
(4) 1.9.28 = 146: AaßO'llOa TtaQa 'HA(Otl äQ~a :rt'tllvWVoQaxov'twv = hypo Med.: ETti äQ~aLoC; oQaxov'twv :rt'tEQW'tWV, ö TtaQ''HALOtl EAaßEV. In (3) as weIl as in (4), it is striking that the Medeiahypothesis and Apollodoros both give details lacking in the text ofthe Euripidean tragedy, viz. the name Glauke and the wingeddragons pulling the chariot of Helios. However, the name Glaukeis known from other mythographical sources (D.S. 4.54.2; schol.E. Med. 19; schol. E. Med. 405; Hyg. F 25; etc.)46), just as the
43) Wagner (cf. supra n. 6) 190; Jouan (cf. supra n. 28) 225; Aelion (cf. supran.28) I 34 with n. 17.
44) Aelion (cf. supra n. 28) I 297-299.45) Cf. Söder (cf. supra n. 6) 88-90; G. Arrigoni, Atalanta e il cinghiale
bianco, Scripta Philologa 1 (1977)38-43; Aelion (cf. supra n. 28) I 315 n.1, 316 n. 8.46) For a more complete list and for diverging traditions, see Roben (cf.
supra n. 14) 871 n.3.
21 Rhein. Mus. f. Philol. 14013-4
322 Marc Huys
dragons (cf. schol. E. Med. 1320: öQUXOVtLVOLt; äQ!-tum); besides,the dragon-chariot of Triptolernos is described in similar terms inseveral sources: cf. ApolIod. 1.5.2 [32]: ÖLepQOV ... Tttl'jvwvöQUXOVtülV; Corno ND 28 (p.54, 1 Lang): TttEQültWV öQUXOVtülV0Xl'j!-tu; Suda Q50 (s. v. 'PUQLUt;): äQ!-tu Tttl'jvwv öQUXOVtülV. Probablythe dragon-chariot was already used in the Medeia of Euripides,although it is not described as such in the text47).
(5) 2.3.1 = 30: :i'tQOt; 'Ioßutl'jv xO!-tLom (Wagner; XO!-tL~ELV Bekker) = hypo Sthen.: :i'tQOt; 'Ioßutl'jv ÖLUXO!-tL~ELV.
(6) 2.4.1 = 34: ZdJt; !-tEtU!-tOQepülSElt; dt; XQuoov XUL ÖLa tfjt;OQoepfjt; dt; tOVt; !luVUl'jt; dOQuElt; XOA:i'tOUt; = hypo Danae: 6 ZEVt; ...XQUOOt; YEVO!-tEVOt; XUL QUElt; ÖLa tOU tfYOUt; dt; tOV XOA:i'tOV tfjt; :i'tuQ-
. SfVOU. The formulation is paralleled not only by Luc. Dial. Mar.12: !lLU XQuoov YEVO!-tEVOV Qufjvm ÖLa tOU oQoepou ... , ÖESU!-tfVl'jV ÖEEXELVl'jV Et; tOV XOA:i'tOV, but also by Men. Sam. 590-1: YEVO!-tEVOt;XQUOOt; 6 ZEVt; EQQUl'j / ÖLa tfYOUt;. This is a strong indication that theterminology used derives from the Euripidean Danae itself48).
(7) 3.5.2 = 36: IlEvSEvt; öE, YEVVl'jSElt; ES 'Ayuul'jt; 'EXLOVL, :i'tuQaKUö!-tou ELAl'jepWt; tilV ßumAELuv = hypo Ba.: IlEvSEvt; öE, 6 tfjt;'Ayuul'jt; :i'tult;, :i'tUQUAUßWV tilv ßUOLAELUV49
).
(8) 3.6.1 = 59: E:i'tL e~ßUt; ... OUV~SQOL~EV = hypo Phoin.:01JV~SQOLOEV ... E:i'tL e~ßUt;. ~1JV~SQOLOE(V) and OUV~SQOL~E(V) arecommon Greek forms for the gathering of military forces: cf.ApolIod. 2.7.7 [156], Xen. Hell. 1.1.15, Paus. 4.17.10, D.S. 19.4.2,D.H. AR 10.14.1, Flav. Jos. AJ 13.337.
(9) 3.6.4 = 65: ÖELXVUOUOllt; öE tilv xQ~vl'jv = hypo Hyps. (P.Oxy. 27.2455, fr. 14, XIV 190: xQ~vl'jv EÖLS[).
(10) 3.6.7 = 73: "AQEL OepuyLOv = hypo Phoin: OepuyLOv "AQEL.~epuyLOv is parallelIed in hypo E. Hec., where it refers to another
47) On the question whether this dragon-chariot was actually shown to theaudience in the produetion of the play by Euripides, see M. P. Cunningham, Medeauno !-ll']xuvTj<;, CPh 49 (1954) 152. It is important to note that Medeia's flight on adragon-chariot is also known from the visual arts, already from Italian vases fromabout 400 B. c.: cf. Robert (cf. supra n.14) 874 n.6; A. Lesky, Medeia, RE 29,Stuttgart, 1931, 62; M. Schmidt, Medeia, LIMC 6, München/Zürich 1992, 391-393,396, nr. 35 ff.
48) Van der Valk (cf. supra n.l) 117-123 thought that Apollodoros borrowed the whole story of Danae and Perseus directly from Pherekydes, but therethe formulation of Danae's conception, if it is verbally quoted in schol. A. R.4.1091, is slightly different: (FGrHist 3 F 10) ZEll<; ... EX TOU ogocpou Xgu04>nugunl.T]ow<; (lEi. Ti bio {mobExnUL 1:4> XOI.JtC{l ...
49) Probably the text of the recently published papyrus hypothesis P. Oxy.60.4017.15-17 (ed. H.M. Cockle, London 1994, 19,22) corresponds verbatim oralmost verbatim with that of the manuscript hypothesis.
Euripides and the 'Tales from Euripides' 323
human sacrifice, viz. that of Polyxena. But in the specific meaningof human sacrifice the term is current both in Euripidean tragedy(cf. Hec. 119, 305: Polyxena; Tr. 747: Astyanax; IA 135, 1200:Iphigeneia) and later (Ant. Lib. 27.2: Iphigeneia, Suda ß 75:Makaria).
(11) 3.6.7 = 73: ea1J1;ov ... EaCjJu~E = hypo Phoin.: eUUtOVunEaCjJu~Ev. The expression may derive from the tragedy ofEuripides itself, where it refers to the self-sacrifice of Menoikeus(cf. I. 1010: aCjJa~U(; E!tUUtOV), but it is also ordinary classical Greekfor a suicide (cf. Thuc. 2.92.3; Xen. An. 8.29).
(12) 3.6.8 = 74: !tovo!tuXOÜGL, xui XtELVOUGLV UAATJAOU<; = hypoPhoin.: !tovo!tuXTJaUVtE<; UAATJAOU<; UVEo..OV. The mutual fratricide byPolyneikes and Eteokles is also described as UVEo..OV UAATJAOU<; inD.S. 4.65.8 and schol. Pi. O. 2.73b, and as !tovo!tuXTJauvta<; unoeuVELV 1mo UnTJAWV in Paus. 9.25.2. However, the use of this phrase isnot limited to the sons of Oidipous, as is apparent from schol. Ar.Pax 284a: !tovo!tuXTJaUVtE<; ö tE K"-EWV xui 0 BguaLöu<; UnTJAOU<;UVELAOV. But whereas the verb !tOVO!tUXELV is used already in E. Ph.1220, UnTJAOU<; UVEo..OV is only found in later sources: see also D.S.5.50.7 and 51.2.
(13) 3.7.1 = 78: KgEwv ... ßUGLAELUV nuguAußWV = hypoPhoin.: KgEWV nugEAußE tljv ßUGLAELuv.The expression nuguAußELV(tljv) ßUGLAELUV occurs in (7) as weIl as in (13), and is current inApoIlodoros (compare 3.8.2 [99] and 3.12.2 [140]) and in manyauthors of the Roman period: e. g. D.S. 4.69.3, 11.68.7, D.H.1.70.3, Flav. Jos. AJ 7.381, 8.2, Plu. Comp. Lyc. et Num. 4.4,Comp. Agis et Cleom. 24.1 5°).
(14) 3.12.5 = 148: EÖO~EV 'Exaßlj xue' 'Ünvou<; (SR, xue' 'ÜnugA) = hypo Alex.: 'Exaßlj<; xue' 'Ünvov otjJEL<;. Compare, for theexpression used by ApoIlodoros, with Dio Cassius, Hist. Rom.(Exc. Salmasiana), Boissevain 3, p. 763: EÖO~E xue' 'Ünvou<;, and forthe hypothesis with Plu. Brutus 41.7: xue' 'Ünvov tMvto<; OtjJLV.Both expressions, then, are ordinary literary Greek in the firstcenturies A.D.
(15) 3.12.5 = 149: ßgECjJO<;, ÖLÖWGLV EXeELVm = hypo Alex.:E]ÖWXEV EXeELVm ßgECjJO<;. The combination of a form of the verbö[ÖW!tL with EXeELVm is customary in descriptions of the act of childexposure: cf. ApoIlod. 3.5.7 [48]: tO yEVVljeeV EXeELVm ÖLÖWGL VO!tEL;
50) For examples from inscriptions see Krenn (cf. supra n. 18) 27, who callsthe expression typical of koine-Greek. This seems dubious, as the earliest examplesdate back to the fourth century B. c.: cf. !soc. Pan. 126, Evag. 35.
324 Mare Huys
hypo E. Melanipp. Sap.: EÖWXE t1i tQoCPep S~[Iv]<;(L; Hyg. F. 187.1:dedit exponendum (a literal translation of a Greek original).The combination ExSElvUL [3QEcpOe; sounds very Euripidean: compare Ph. 25, where it refers in a prologue to the abandonmentof Oidipus.
(16) 3.12.5 = 150: we; [ÖtüV Ttalöu EtQECPEV, OV0I-lCWUe; IIuQLV= hypo Alex.: ESESQE'tjJEV ULOV ... [ II]uQlv JtQoauyoQEvaUe;.Compare with D.5. 4.33.11: we; [ÖtüV ULOV EtQECPE, JtQoauyoQEvaue; TtlAEcpOV ... , on the discovery of the abandoned Telephosby the herdsman Korythos.
(17) 3.13.8 = 174: xQv'tjJuau EaSijn yUVULXEL<;X ... AUX0l-ltlÖElJtUQESEtO = hypo Skyrioi (PSI 1286, col. 11, 11. 15-16): xOQT)e;EaSijt[l xQv'tjJuau JtUQESE]/tO AUXOl-ltlÖEl51). The editio princepshad xOQT)e; EaSijt[U MI, JtUQESE]/tO AUX0l-ltlÖEl. I accept here thesupplements proposed by Luppe. At least JtUQESE]/tO is very attractive: "to give a person in charge to" is an ordinary meaningof the verb (cf. LSJ s. v. B 2) and we have it again in schol. T326: JtQoe; Auxol-ltlÖT)V tOV [3U<JlAEU JtUQE'l'tEto tOV 'AXlnEU.
(18) Epit. I 19: euutTjv uVtlQtT)aE = hypo Hipp.: UUtTjv öEuvtlQtT)aE, The expression euutTjv uvtlQtT)aE(v) regularly refers tosuicide by ha~ging in contemporary literature: cf. Flav. Jos. AJ7.229; Ant. Llb. Met. 13.4; Plu. Lucull. 18.6; Apollod. 1.9.27[143],2.7.7 [159],3.5.9 [56], 3.12.6 [155], 3.14.7 [192].
(19) Epit. 111 5: 'EAEVT)V I-lEV uJto 'EQl-loÜ ... X0I-lWSijVULxAuJtElauv de; A[yuJttoV XUL öoSEi:auv IIQwtEi: tep [3U<JlAEi: tWVALyUJttLWV CPUAUttElV = hypo Hel.: XAE'tjJUe; YUQ UlJtTjV 6 'EQI-lije;... IIQwtEI tep [3u<JlAEI tije; ALyVJttOU CPUAUttElV JtUQEÖWXE. The expression EÖWXE CPUAattElV is common Greek (cf. Isoc. 21.2) andcorresponds with E. Hel. 910: EÖWXE a0~Elv. Therefore, and because of differences of content (cf. supra), Kannicht denied anydirect connection between this hypothesis and the Epitome52).
(20) Epit. V 23: 'AatuuvuXta uJtO tWV JtvQYwv EQQl'tjJUV,IIoAuSEvT)V ÖE EJtL tep 'AXlAAEWe; tucp<p xatEacpusuv = hypo Tr.:IIoAuSEVT)V ... I-lEV o'Üv EJtL tije; tOÜ 'AXlnEWe; tacpije; Eacpusuv,'A<JtuuVUXta ÖE uJtO twV tElXWV EQQl'tjJUV. This bipartite sentence53) is parallelled by D. Chr. 11.153: 'Aat1JuvuxtU I-lEV ...
51) Cf. C.Gallavotti, Nuove Hypotheseis di drammi euripidei, RFIC 11(1933) 177-188; W. Luppe, Die Hypothesis zu Euripides' 'Skyrioi', Anagennesis 2(1982) 268-269.
52) Cf. Kannieht (cf. supra n. 37) II 241.53) On this passage of the Library, whieh probably eorresponds with a
version by Arktinos or Lesehes, see Wagner (cf. supra n. 6) 243-245, 253-254.
Euripides and the 'Tales from Euripides' 325
UVEf..ELV Q('ljJuV'ta~ uno .ou 'E(XO'lJ~... , Ilof..'lJl;tvTlv öt: nUQ8evov unompuLLELv futL .u<pq>.
So, some correspondences quoted above may be simply accidental, the result of the treatment of the same legendary event in amythographic fashion and with the vocabulary characteristic ofthat period, whereas others, such as the use of the verb 1i0VOliUXELVfor the duel between Polyneikes and Eteokles, may be due to acommon indirect source, viz. the drama of Euripides itself54). Examples pointing with a measure of certainty to a closer connectionbetween the Library and the Tales are particularly rare: in myopinion the comparison between hypo Alk. and the correspondingsequence from the Library yields the most striking verbal parallel.Söder concluded that the correspondences between the two texts"verraten die Verfasser von Kompendien als Benützer solcher Hypotheseis". But two lines further, Apollodoros already divergesfrom the Euripidean version, writing that Kore sent Alkestis backup from the Underworld. Euripides is meant here with the periphrastic plural EVLOL, who opted for the variant of Herakles' bringing her back55). Once again, several sources have been intermingled. Still, this case proves that the Library contains sequences thatgo back at least to material used also by the author of the Tales[rom Euripides. Although it is my personal view that these Tales ofthe type found onlapyrus are not from the hand of Dikaiarchoshirnself, but shoul probably be dated to the later Hellenistic ageor the first century A. D., they seem to be rooted in Hellenisticscholarship - the introductory formula (tide 0-0 uQXTJ, the first lineof the play, ~ ö' vn68E(JL~) is indeed close to that of the Callimachean Il(vuxE~56). It must be stressed also that there seems to havebeen more than one type or collection of the Tales [rom Euripides:we have one papyrus with hypotheses arranged not in an alphabetical order but according to another principle (lFAO, PSP. 248)57),and some important examples of hypotheses preserved in Byzantine sources together with part of the prologue of the play con-
54) According to Bethe (cf. supra n.6) 86, the mythographic manual onwhich both Apollodoros and Diodoros depended here closely followed the text ofthe Phoinissai itself.
55) Cf. Söder (cf. supra n. 6) 132.56) Cf. R. Pfeiffer, History of Classical Scholarship from the Beginnings to
the End of the Hellenistic Age, Oxford 1968, 129, 195.57) Cf. M. Papathomopoulos, Un argument sur papyrus de la Medee
d'Euripide, Recherehes de papyrologie 3 (1964) 37-47: in this edition the hypothesis of the Medeia is preceded by a hypothesis of the Peliades: so the summaries seem to have been arranged according to their legendary content.
326 Mare Huys
cerned (Stheneboia, Melanippe Sophe and perhaps Peirithous)58) it has been argued by D. F. Sutton that aseparate collection of suchhypotheses may have existed in Byzantine times, consulted also byJoannes Tzetzes59). Finally there seems to have been a shorter orabridged type of hypotheses, of which some have been preservedin the manuscripts of Euripides. For example, as was already observed by Zuntz60), the manuscript hypothesis of the Alcestis is ofa much shorter and more concentrated type than the usual Tales,and the manuscript hypothesis of the M edeia as well as the secondhalf of that of the Helen exhibit the same characteristics. Luppeand other authors have argued that the manuscript hYfothesis ofthe Alkestis is a later abridgement of the hypothesis 0 the Tales,known from fragment P.Oxy. 27.245761
). The arguments used,however, do not rule out the possibility that hypotheses of bothtypes circulated from the same date onwards or that the author ofthe Tales had used for convenience portions of an existing hypothesis.
It seems likely, then, that one of these collections of hypotheses was a source of the Library, but it cannot be proved thatthe mythographer hirnself excerpted this source. As to the possibility that he would have consulted the text of Euripides, thisseems very improbable, and even for the prologues, which mighthave been included in a collection of hypotheses, the argumentsused do not convince me. Apollodoros was indebted, directly ormore probably indirectly, perhaps through the intermediary of amythographic manual, to Alexandrian scholarship, hypothesesand learned commentaries. This has been amply demonstrated forsome specific cases, e. g. for Apollodoros' treatment of the legendof Medeia's infanticide, by scholars such as Robert and Söder62).
Thus, Bethe's warning remains valid: "mihi persuasum habeo,saepissime praecipue in bibliotheca et in Hygini fabulis multorumfontium frustula ita esse confusa et contaminata, ut omnino nonpossint distingui", as does the conclusion of Söder: "Das mannigfach verschlungene und verstrickte Gewebe der mythographischen
58) It is not certain that the quotation preserved from the last play actuallycomes from the prologue: cf. D. F. Sutton, Two Lost Plays of Euripides, NewYork 1987, 33-34, 95-96.
59) Cf. D. F. Sutton, Evidence for Lost Dramatic Hypotheses, GRBS 29(1988) 90-92.
60) Zuntz (cf. supra n.l) 144-145.61) W. Luppe, Die Hypotheseis zu Euripides' 'Alkestis' und 'Aiolos'. P.
Oxy. 2457, Philologus 126 (1982) 10-18; other authors are quoted on p.l0 n.2.62) Robert (cf. supra n. 1) 83; Söder (cf. supra n. 6) 160-162.
Euripides and the 'Tales from Euripides' 327
Literatur im einzelnen entwirren zu wollen, erscheint so aussichts10s."63) Too often this has been forgotten when using the Libraryfor the reconstruction of lost tragedies. To consider a passage fromthe Library as an argument from a Euripidean tragedy and a fabulaof Hyginus as the argument of the Sophoklean treatment of thesame legend or vice versa is too simplistic64). Sometimes, indeed,these passages do contain material from tragic hypotheses andscholarly commentaries on tragic texts, but their contamination byother sources is often impossible to disentangle. My conclusion,then, accords with that reached recently by P. Dräger65) after anextensive comparison of BibI. 1.9.16 [107-109] and PherekydesFGrHist 3 F 105, viz. that the Library does not depend directly onthe original source but constitutes a late phase of mythographicalactivity going back to Hellenistic scholarship. The sometimes abstruse elucubrations of Van der Valk and the somewhat naive adoration of Carriere and Massonie have attributed to our booklet ascholarly originality to which it can lay no claim.
Leuven Marc Huys
63) Bethe (cf. supra n. 6) 99; Söder (cf. supra n. 6) 169.64) Cf. supra n. 39.65) P. Dräger, Argo Pasimelousa. Der Argonautenmythos in der griechi
schen und römischen Literatur, I: Theos Aitios (Palingenesia 43), Stuttgart 1993,42-63, esp. 62. However, I do not follow this author (97 n. 281) in his criticism onSöder; his thesis that the authors of the tragic hypotheses used the Library insteadof the other way round remains unexplained.